40 Year Pick Anniversary

[Fred Waltman]

Just realized that as of last month I have been working with Pick like systems for 40 years.

It was 40 years ago last month that I walked into the offices of the Microdata dealer in Toronto and they handed me a think book with a green cover and a smaller one with a rainbow cover and said "Read this."

Time flies when you are having fun...

[Dawn Wolthuis]

Congratulations, Fred! It sure sounds like we would collectively benefit from a few more stories. If you prefer to say them instead of writing them, shoot me an email off list (dwolt at tincat-group dot com).  

Cheers! --dawn

Typed on a mobile keyboard

--
You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com
To unsubscribe, email to: mvdbms+un...@googlegroups.com
For more options, visit http://groups.google.com/group/mvdbms

[Steve Trimble]

yep, 38 years and counting for me sir Fred. hope you're doing well. fun times ahead

Steve Trimble

Computerized Data Mgmt Inc

(501) 615-8674 office

(501) 772-3450 cell / text

--

[geneb]

On Wed, 15 Jun 2016, Steve Trimble wrote:

> yep, 38 years and counting for me sir Fred. hope you're doing well. fun
> times ahead
>

30 here. I must be the youngster in the crowd. :)

g.

--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.

ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://scarlet.deltasoft.com - Get it _today_!

[Glen Batchelor]

No I am.. Always the youngest at Spectrum too. LOL

[geneb]

On Wed, 15 Jun 2016, Glen Batchelor wrote:

> No I am.. Always the youngest at Spectrum too. LOL
>

So if I ever manage to make it out to one, should I bring you a rattle? :)

[Glen Batchelor]

Nah, teething ring will work

[geneb]

On Wed, 15 Jun 2016, Glen Batchelor wrote:

> Nah, teething ring will work
>

I'll print you one out of Semiflex. :)

[Peter McMurray]

Hi Fred

40 for me too. Reality 2.4 with a massive 10 megabyte hard disk and 16Kb of core memory. At the time I was considered a bit of an NCR assembler guru, the shock to the system was enormous. Frank Broos of AWA used to sell them off a whiteboard using ENGLISH . I do like the "think book" :-)

[Fred Waltman]

Doing well, Steve... Hope your are as well.  A long time since that Spectrum back in the early 80s...

Fred

[jim stephens]

I started @ Microdata 40 years ago this month.  Sometime that fall they had the "Sweet smell of success" event, and the ad series with roses, celebrating 1000 systems shipped. 

Lots of fun.  They tried almost every wrong way of doing things that a company could try, and survived quite a while in spite of it.  Once we all had been thru that, there weren't a lot of surprises left companies could try.

Great working "inside" the system for as long as that afforded me the opportunity to do so.

Jim Stephens

[Dawn Wolthuis]

Thanks for the story, Jim! Congrats on your 40th too.  --dawn

Typed on a mobile keyboard

--

[Art Martz]

I had to stop and figure mine out:

(DATE()-ICONV('10/08/1978','D'))/365  = 37.7178

(I know it was *some* time in October....

[Chris Long]

I guess I’m the young ‘un of the group.

 

Started in Reality on Julian date 11688, 12/31/99.

 

Chris Long

RA Services IT

[Will Johnson]

7.6.1983
internal date 5666

33 years and counting

[Peter McMurray]

Since we have got into the date does anyone remember why

:date 1

January 1, 1968; Monday   internal: 1   julian: 1

Dick was simply after the first day of the year as the first day of the week.

A big thank you to the writers of PICK BASIC for what is still the very best date routine that I have ever come across in any product.

[John Lorentz]

On Fri, Jun 17, 2016 at 6:44 AM, Art Martz <artsm...@gmail.com> wrote:

I had to stop and figure mine out:

(DATE()-ICONV('10/08/1978','D'))/365  = 37.7178

(I know it was *some* time in October....

Which is 15 days before I started working on Pick systems, at ADP (10/23/78).

John

[Joe Fisher]

As the longest MVer so far on this thread, I must weigh in, and heavily. In 1974 Microdata launched a large trade press campaign to advertise their new  Reality system, especially English. I was using an IBM System 3 at a service bureau then and some of Microdata's ads compared Reality with the System 3. After a month or so of reading the ads' preposterous claims, I finally called Microdata.

The national sales manager answered the phone, we talked a while and then he gave me a number to call of a Microdata dealer in Dallas. After several phone calls I agreed to go to Dallas to see their Reality 2.1 (I think) 20 MB system. They claimed it was the 4th system that Microdata had shipped.

The dealer's company was in the insurance industry and had been using some RPG programs which they thought they might want to buy and convert to Reality. I told them I'd done some RPG work and had a program in 128 column paper cards that we could convert to mag tape and test. So in October of '74 I went to Dallas and met a contractor to Microdata called Pantech that was developing an RPG compiler for Reality. It was two brothers named Gearheart; Ed had come to Dallas to meet me while his brother remained in CA.

"Blown Away" is the only description of my reaction! My program did statistical analysis of testing for a professional association and used at least Square Root and Table Lookup. Sitting at a video terminal, Ed made changes to the Assembler code of their Beta version compiler, explaining every step. He called his brother several times so they could agree on and implement fixes. Wow! In real time too!

I had never even imagined a table driven assembler that was extendable in its own language. Years later I developed tables and an extension to assemble Zilog Z80 microcode on Reality for a project. The extension resolved the "high byte low" to "high byte high" addressing conflict between the Reality firmware and the Z80 architecture.

Anyway, Ed finally pared my program down to something that he could compile and ran it. And explained the Spooler design and control as he did a print out. Unbelievable. I don't think the RPG compiler ever appeared as a salable product.

Attached is a snapshot of the top of the pin board in my study. At the top is the rack header plate from a Reality system, long ago scrapped. The ball caps are (L-R) openSUSE, ubuntu, and Pick, never seen another one like it. To its right is a page from a calendar that explains the (Pope) Gregorian calendar, not adopted in the British world in 1752.

And lastly (thank goodness) since Peter McMurray mentioned the Pick calendar, I agree that it is really the best. It rationalizes the data of (any) calendar instead of dealing with various external representations of that integer data. Implement the Jewish or Lunar calendar in Assembler as an xCONV with its own unique "conversion code" and you'll see.

BUT, Peter, I believe Pick's calendar has a flaw: it has a February 29, 1900! There was no such day! In the Gregorian calendar, years evenly divisible by 100 are not leap years, unless they're divisible by 400. Which 2000 was. I doubt that has ever been an issue in MV programming. Try it.

Thanks to everyone for your comments and anecdotes, I love Pick system history.

[Peter McMurray]

Hi Joe

I must admit I had never tried the 1900 trick. I assume someone fixed it long ago as current D3 is correct 

:date feb 29 1900

Invalid date!

However Reality was not alone in this I remember EXCEL had this fault. Those  outside the insurance industry probably never even knew but it could cause grief in valuation calculations. My math has long since faded in this area but if I remember correctly it was  Newton–Raphson method. 

An area that did catch me out mightily was the default century starting in 1930. On my first payroll program I set a data entry test on the birthdate that a person had to be born already. A bemused data entry lady came to me saying it would not let her enter Fred. She was happily banging in day dot month (this was 1978) and Fred had been with us since 1925. OOPS!

Also fixed in D3

:date 29-1

January 29, 2016; Friday   internal: 17561   julian: 29

Paper cards brings back memories - please don't drop the box. I also remember punching paper tape for NCR programs; plugging wires into a "breadboard" for a Casio I think, drawing chinagraph marks on a massive sheet of plastic for an NCR 299, altering running programs in hex in memory and saving the result back to the master tape NCR 399 - please remember to alter the source code or later understanding became interesting but that took three hours to compile and a direct fix worked now.

I possibly have a couple of years on most here as at 75 I am reducing my daily programming and have started Music Theory so I can now blow my mind with Coltrane Progressions instead of stack overflows.

[Chris Kusmierz]

I started programming  in FORTRAN on an IBM 360 with punch cards in the 70s. Never even heard of MV until 2001. Been maintaining a PICK application since June of 2007. Still just a noob.

Chris Kusmierz

[Rob Allen]

In 1978 I graduated from college and got a job at a service bureau supporting clients on an IBM 360 with punch cards. Early in 1979 the company announced that they had developed a minicomputer version of their product, and they asked for volunteers from the customer service group to learn the new system. I raised my hand, and was thus introduced to Reality. 

Later on, at Sequoia, I sat next to Tim Holland as he worked out the algorithm to correct the "Feb. 29, 1900" problem. 

--

Rob Allen

[Will Johnson]

I got my first Pick job because of https://en.wikipedia.org/wiki/Colossal_Cave_Adventure

One of the questions was "How many hours have you spent in front of a CRT" and I said probably a hundred or more
Because I was addicted to this game
And so I was hired

[Dawn Wolthuis]

I'm a late-comer to PICK, but started in computing with a job writing COBOL (and a bit of FORTRAN) on a Pr1me 300 computer in 1977. Soon I had "my own" P400 for writing COBOL with MIDAS files financial systems while my boss worked on warehouse applications on his P400. I sent my first inter-machine email in 1977 over PrimeNet.

During about 7 years on big iron, writing CICS COBOL apps with both VSAM and IMS DL/I. I saw a lot of other tools on MVS, VM, and DOS including RPG, PL/1, SAS, JCL, SPF/ISPF, Datatrieve, Intellect, Oracle, Dbase, R:Base, and DB2. I also took a look at Windows 1.0 and an early Mac. 

In 1988, when I started in my first management position at a place running Pr1me computers, I saw Prime Information for the first time. I recall my exact words were "that is not a database." There was an application written internally for our helpdesk on PI. I mused in an early meeting that it sure would be nice if we could collect information about this or that. Later I suggested we schedule that development work. I was told "Oh, I put that in when you said it would be nice -- it's there." "In production?" I asked. "Yes."

I knew the task would require database design with data dictionary additions and changes to existing programs. It was not just an add-on, but integral to some of the application The developer told me it took her less than 30 minutes to go from me saying it would be nice to having it in production. I found this to be both absurd and delightful. It was written in SIMPLE on PI. (Anyone remember SIMPLE?)

Although I now consider hash tables within the MV space to be "real-enough databases" now, like many others here, I do not limit myself to PICK. We can run MV on much more typical industry platforms, preserving the elegance of the NoSQL document-database data model while still working with tools such as those from Microsoft. There is plenty of R&D happening in that arena right now.

Oops, we were talking history and I moved to futures, so I'll end there.  Cheers!  --dawn

--

You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com
To unsubscribe, email to: mvdbms+un...@googlegroups.com
For more options, visit http://groups.google.com/group/mvdbms

--

Dawn M. Wolthuis

Take and give some delight today

[Dawn Wolthuis]

A good read, Jon. Thanks!

--dawn

On Mon, Jun 20, 2016 at 4:54 PM, jes <j...@jes.com> wrote:

I was born a poor black child in Mississippi.

Too poor to afford my own computer, I built my first one out of a cardboard box, bailing wire and wood chips. It was powered and cooled by a nearby stream.

Still too poor to afford new punch cards when I started COBOL at a State School, we sat around the kitchen table in the evening recycling punched cards by filling in the holes.

Paying back the teachers who scolded me for always using block print instead of cursive, I excelled at an RPG2 course provided by IBM in Riverside, California in 1976. 

Based on my programming experience with these two languages, I became a singer/guitarist in a duet called “Prime Time” in 1978.

That same year, I started a business providing software for the IBM 5110 - the precursor to the PC (Model 5150) and quickly was overwhelmed with success from IBM reps tasked with selling as many of these as fast as possible, before customers realized they just paid $26,000 for a computer with twin 8-inch, 1.2mb floppies, 32K of main memory and a 4-inch screen. My favorite thing to do with this was flipping the front panel switch that toggled it from Business BASIC to PL1. Until we discovered ASCII porn.

This resulted in a need for more programmers, so I ran an ad looking for those who could code in BASIC, requiring them to submit samples of their code over resumes.

One kid submitted a mailing label update routine written in something called “DATA/BASIC”. It didn’t look like anything I had ever seen, so I asked to see the machine it was written on. He invited me down to a company who wrote apps for the Medical industry for a tour. 

They had a glass-walled computer room with a bunch of huge REALITY machines, but the thing that impressed me the most was the acolytes tending to sacrificing half-inch tapes to these monsters all got to wear White Lab jackets. How cool was that?

The CFO somehow got wind that there was a visiting programmer in the house who not only knew how to code but how debits and credits affected balance sheets. I soon figured out this was me. He did too, and offered me a job on the spot to fix their in-house accounting apps.

A year or so later, a dude from Microdata Tech Support came to town to do a weekend training session. We hit it off and he suggested I submit a resume to Microdata and two weeks later, I had given up my apartment, gave the business to one of my College Professors and found myself in the Microdata Tech Support group on McGaw Ave. in Irvine. That was 1979.

I still had not even heard of Pick by the time I was sent on my first spy mission to an Ultimate dealer (That Bill (something) dude who always creepily carried a Bible around.)

Many happy days (2) were spent at Microdata. 

It was especially exciting to be sent to prestigious user sites experiencing REALITY issues no one else could figure out, without a lick of experience or training in the problems they were facing. If NASA had worked this way, we would never have made it to Fort Lauderdale, much less the Moon.

In 1981, I got my hands on an HP3000 Pocket Guide and had an epiphany,

I left Microdata and in the course of 30 days, wrote and produced 200 copies of the REALITY Pocket Guide which I toted to the Micru Con in NYC and sold every last one at $50 each. This almost covered my hotel room bill and would have had I not ordered a bagel and OJ from room service.

Pick got wind of the RPG and invited me over to discuss making one for “his” version of MD512, which he cleverly called “Pick”.

Never one to miss an opp, I proceeded to retool  the RPG into the Pick Pocket Guide (still a great name if you ask me), which I estimated at taking 6 weeks. I’ve never been good at estimating. 

Plus, I had not anticipated Henry Eggers, who took a very Paternal role in the platform and the words used to describe it, That 6 weeks turned into 6 months, but it was an unanticipated mentorship under a legend, so well worth it.

For those who know Henry, you would understand the exponential bump in production time.

You could ask Henry a question like “Henry, is it true that the R option redisplays the line of text?” and get a 3-hour dissertation on the ins and outs of why and how RBIT got set.

Little did Henry know that he could just as well have been explaining this to his Golden Retriever, as I had no idea what registers were, much less ever having even seen Assembler languages.

After Henry either ran out of cigarettes or we hit the three hour mark, I would wait for a moment of silence and ask “so was that a yes?”.

All these years later, we still laugh about RBIT.

Unfortunately, it’s at times like funerals and standing in a bank teller line that this happens.

The rest, as they say, is history.

--
You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com
To unsubscribe, email to: mvdbms+un...@googlegroups.com
For more options, visit http://groups.google.com/group/mvdbms

[Roland Ruiz]

Peter - It was me that added the leap year logic, circa 1999. This was 11 years after walking into Pick Systems for the first time.

Long live the Pick!

Cheers,

Rolando

[HB Electronics]

Peter - It was me that added the leap year logic, circa 1999. This was 11 years after walking into Pick Systems for the first time.

Long live the Pick!

Cheers,

Rolando

On Sun, Jun 19, 2016 at 4:19 PM, Peter McMurray <pgmcm...@gmail.com> wrote:

[Roland Ruiz]

Peter - It was me that added the leap year logic, circa 1999. This was 11 years after walking into Pick Systems for the first time.

Long live the Pick!

Cheers,

Rolando

On Sun, Jun 19, 2016 at 4:19 PM, Peter McMurray <pgmcm...@gmail.com> wrote:

[Will Johnson]

There is a copy of the SIMPLE User's Guide at the Computer History Museum in Mountain View

The below blurb makes it sound like it was a 4GL ??

"This guide explains how to use

Prime INFORMATION SIMPLE to create 
files, describe dictionaries, enter data on 
formatted screens, and perform inquiries 
and reports on that data. It provides the 
basic terminology needed to create, access, 
and maintain SIMPLE files. It illustrates 
the menu that provides the interface 
between the user and the Prime 
INFORMATION System, and details each 
menu option. It also describes the SIMPLE 
functions that are available outside of the 
menu options. 

As revised for SIMPLE Release 6.0, the 
book documents the enhanced SIMPLE 
product, which includes a separate menu 
for utility options. Prime INFORMATION 
Release 5.4 is required for this version of 
SIMPLE. "

[Richard Wilson]

I could use a copy of that as I still have one Prime client using simple

Rich

> --
> You received this message because you are subscribed to
> the "Pick and MultiValue Databases" group.
> To post, email to: mvd...@googlegroups.com
> To unsubscribe, email to: mvdbms+un...@googlegroups.com
> For more options, visit http://groups.google.com/group/mvdbms

--
Richard A Wilson
Lakeside Systems
Smithfield, RI, USA
Voice 401-231-3959
Fax 206-202-2064

[Wols Lists]

On 21/06/16 19:49, 'Will Johnson' via Pick and MultiValue Databases wrote:
> There is a copy of the SIMPLE User's Guide at the Computer History
> Museum in Mountain View
>
> The below blurb makes it sound like it was a 4GL ??

As I understand a 4GL, it was. Not as nice as &PACE, but it was pretty
decent. IIRC a lot of the code I worked with was originally generated by
SIMPLE. Snag is, a lot of it was then hand-modified, and once you did
that, you were stuck with it. (And of course, like any 4GL, it tended to
be pretty verbose code.)

Actually, I think the source code for SIMPLE is floating around
somewhere - I might even have a copy myself somewhere on my hard drive
or CD backups ...

Cheers,
Wol

[Dawn Wolthuis]

Yes, that's it! Yes, it was a 4GL tool for screens with most of the work in typical InfoBASIC, dicts, and paragraphs.

Thanks.  --dawn

--

You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com
To unsubscribe, email to: mvdbms+un...@googlegroups.com
For more options, visit http://groups.google.com/group/mvdbms

--

[Will Johnson]

That's interesting.  The Prime Information section of the Wikipedia Prime page here

https://en.wikipedia.org/wiki/Prime_Computer#Prime_Information

says that Prime is obsolete.

[Dawn Wolthuis]

Well, I feel a stab in my chest. 

I set the switches, booted up "my own" P300 (then P400), threaded the tapes, swapped out "portable disk packs" and talked with Pr1me engineers to fix bugs in the COBOL compiler. 

I soldered pins for RS232 connectors, crawled through the ceilings to drop wires, and wrote up step-by-step instructions for end-users.

I sent email over PrimeNet. I hacked into a Pr1me in the Engineering department at Michigan State when the Math grad students were not given accounts (they had a custom login and I knew I didn't know how to write one of those without permitting <ESC> to exit to the OS) in order to show it to my colleague who knew the PDP-11. Oh, because an engineering student was trying to police me, I created a directory called .NULL. so when he tried to figure out where I was, it appeared I was nowhere.

I wrote a Pr1me Runoff (written for the Pr1me by Martin Phillips, IIRC) book on COBOL for a college course I taught in 1981. I knew how to do fancy things with ED, work with CLIST and MIDAS files and make changes to the Fortran code that the OS was (originally) written in. My first management position was for a department where we eventually had a fleet of 7 Pr1me computers in the early '90's. The most expensive "server" (aka mini-computer) I ever budgeted and signed for was a Pr1me 6650. 

Prime is obsolete already? It can't be! That would make ME obsolete, perhaps? Smiles. Moving forward. --dawn

--

You received this message because you are subscribed to
the "Pick and MultiValue Databases" group.
To post, email to: mvd...@googlegroups.com
To unsubscribe, email to: mvdbms+un...@googlegroups.com
For more options, visit http://groups.google.com/group/mvdbms

[Kevin King]

Thus proving that Dawn was the real, inaugural "Mr. Robot".

-K

[Ed Clark]

Just keep in mind: you’re never dead if someone remembers you (which can either be a classical cultural reference, or one from as recent as last night).

I remember twiddling switches on and feeding various tape and card media into DEC PDP8s and PDP10s and IBM System 34s (and using key-to-disk on 8 inch floppies for the 34), and snapping together RS232 connectors (never soldered) for ADDS Mentor systems and a big, indefatigable Stratus running Pick OA. Those things are all gone, obsolete. But not forgotten. You can even run emulators of them on your handheld (or wrist-mounted)  devices. Emulators aren’t the same as the original, of course: there’s no emulated experience for opening up a Commodore 64 every couple of months and cleaning the key contacts with an eraser so that you can type without a hammer. But the experience isn’t *forgotten*.

[Peter McMurray]

I never met SIMPLE but it sounds remarkably similar to the program that my partner Stuart Evans wrote in 1976 for the AWA VTE6 as a precursor for us to write the first genuine Pick 4GL Pobol..

Stuart's little joke, as a senior member of the UK Cobol committee he chose the Welsh word for people. Stuart knew Lionel Singer well and salvaged Lionel's initial Wang disaster for him.

As for Prime being obsolete; I was working on the Trust bank code in 1999. They had hooked two Primes together to cope with the load. I was the person who had to sort out why the replacement HP machine "fell off a cliff" every morning when the credit card updates were run. The major accounting firm responsible had specified RAID 50 which was totally unsuitable for the tens of thousands of transactions being thrown at it in a second or two. The RAID 10 replacement upped the price by $20,000 - not a popular management discussion. The Prime solution had worked fine.

Dawn can lay claim to being a serious Black Hat hacker - I love the .NULL fix. I can remember hunting down null keys in Pick on the Reality using the magnificent DUMP GX command sadly missing from FSI. However I simply copy a file I need to check into the VME when doing UNICODE testing and it still works fine. As for dodgy nulls and control characters in data I simply tightened the standard data entry routine and operator finger problems vanished over 30 years ago.

[jim stephens]

"Just keep in mind: you’re never dead if someone remembers you (which can either be a classical cultural reference, or one from as recent as last night)."

*JAT
*SCOTT
*AULER
*HVE

*CHANDRU (rare in the code)

don't recall others.

[Ross Ferris]

I remember Simple. I was contracting at the Maritime Services Board of NSW & because it was taking soooo loooong to compile screens (remember) I sent them off as phantoms so I could be more productive - I got a call within an hour and asked not to do that - had about a dozen compilations going & performance across the state had users up in arms. A great introduction to Government .... work slower :-)

[Peter Gonzalez]

30 years for me! I started as an entry level programmer in Soulth Florida.

-Petrr G

[JJCSR]

I just recently began the 55^th year in this field of endeavor – be it known as EDP, MIS, IT, or whatever moniker our bread-winning years have held.

 

I spent 6 months, from Oct. 1961 to April 1962, in a class that taught operations and board-wiring for EDP (Electronic Data Processing),  IBM equipment - 402, 407, 514 etc., etc.    Landed a job in May of '62 in the EDP department of E.F.Macdonald's (EFM) (Sales-incentive processing) in Dayton, Ohio.   EFM employed a staff of 40-50 keypunch and verifier operators (the ABSOLUTE prettiest of which became my wife - now of 50+years -, and mother of our 5 children), with some 10-12 operations personnel.   Much of my time on the job early on at EFM was spent sorting, literally, 10's-of-thousands IBM 5081 punched cards, subsequently processing through IBM 407's, summary-punching to 514 card-reproducers.    In mid-1963, EFM took delivery of one of the first IBM 1401 systems in the area.   With that system, I began my years of computer-programming, using AUTOCODER.   Three-four years later, I joined a company that was starting up their MIS department to service their 30 retail lumber and building product stores, as well as three wholesale distributor locations.

 

By 1969, I was managing the MIS group, and by spring, 1983, had seen growth from the initial IBM 407, through IBM 1440, to System 360/25, to 360/30, to 360/40, and finally 370/138.   At that point in time, I saw an ad for someone with retail/wholesale experience in MIS, to startup a department for a sporting goods outlet on the east coast.   I won the job, and by 1984 had selected two PICK-Based software "packages", one for wholesale operations, and one for retail.   The first system installed was a GA 5500, later to be joined by a GA 5820.    In May of 1984, I ventured to JES and Associates, Laguna Beach, Ca. for my "Intro to Pick" class.   By October, 1984, I was sitting in a newly-completed computer room, with the GA 5500 and some dumb terminals.   The 9-track tape drive allowed me to offload the wholesale package, print out some programs, and, having some 20+ years of programming behind me, I started teaching myself PICK/Basic, and re-writing some of the programs for our personal needs.

 

Today, some 32 years away from introduction to the MV world, yet over 54 years removed from having kicked off this IT career, I still get a lot of enjoyment out of my work.   We have seen GA Pick, to Data General PICK on DG/UX, to Advanced Pick on NT, to D3 on Windows, and to present REALITY on VMWare, running Failsafe redundant systems.

On Wednesday, June 15, 2016 at 9:13:56 AM UTC-4, Fred Waltman wrote:

Just realized that as of last month I have been working with Pick like systems for 40 years.

It was 40 years ago last month that I walked into the offices of the Microdata dealer in Toronto and they handed me a think book with a green cover and a smaller one with a rainbow cover and said "Read this."

Time flies when you are having fun...

[Peter McMurray]

Hi

Autocoder. Now I well remember my partner thinking that was the death of programming. He was well known for using assembler and always running from zero - debugging delight.

What worries me is where are the youngsters?

[Fred Waltman]

While it has been 40 years on Pick, my experience with a Microdata 1600 predates that by a year. While studying at University of Waterloo we had a Microdata with a Writable Control Store -- that is the "firmware" was software. It was hooked up to a slot car set and it was used to teach real time programming.

Actually, now that I think of it we also had a Photon phototypesetter which was built around a Microdata 1600 as well.

Most of my programming back then was in B (C was this new thing that only ran on PHP 11s).  In one class we worked on what they were calling a "portable version of B/C" -- since this was Canada it was called "eh"

When C comes up in a conversation, I love to tell people I am so old I programmed in B -- mostly it gets blank stares. Us old farts have a different sense of humor I guess :)

[Peter McMurray]

Love it Fred. Being Canada I would have expected a French linguistic derivation. Lo and behold you come up with a German one. Yes in the Diatonic Scale the Germans use H for the second letter B it comes from Pythagorean intervals. http://www.tonalsoft.com/enc/g/german-h.aspx gives a delightful explanation.

[Joe Fisher]

Fred,

In ~1986 I really needed a 1600 Writable Control Store board, but couldn't find one. A client of mine had 3 Microdata Royale (3.0 firmware running on a 1600) systems and did their own hardware maintenance, with my help. Thus we had the Production, the Backup and the "Hanger Queen" systems. Then one of the firmware boards failed and Microdata wanted ~$3K+ to fix it. Managment wouldn't pay: "machines are too old", they said.

So we put 2 General Purpose I/O (8-bit GPIO) boards in the Backup and I built cables to attach those boards to the front panel of the Hanger Queen's 16 indicator light drivers and some control switch buffers. Then I wrote Assembler code to operate the STEP switch on Hanger Queen's front panel to go through the Royale microcode, input the display lights and put it all into an MV item in a file. That gave us the "bad" board's code.

Late one weekend night, we shutdown Production and moved its firmware board into Hanger Queen and did the same thing, giving us a "good" board's code. Comparison of the two boards' code showed only one of the board's wire ROM chip had failed. It cost $0.79 for a new part which we burned on a borrowed Kontron PROM burner, having downloaded the new chip's "good" code. After replacing the bad chip the board ran like a bandit!

My client's managment never questioned my invoices, even tho' they probably paid more than $3K! A Writable Control Store board would have been easier, I bet. I knew about such things, but most non-Reality 1600s were long gone by then. Less than 2 years later, they went to an 80286 based co-processor board running Ultimate installed into a DEC microVAX II. That was a good system too.

[Will Johnson]

I thought my microVAX running Ultimate at the Life Extension Institute was the only one!
We ran out of back-plane ports and they wanted at least 20K just to add Ethernet, so instead I hooked up a cheap ADDS and made the two systems talk to each other for seven more inquiry only ports

[frosty]

Possibly Chandru was using his initials, CM. Not to be confused with Carl Margolis or Cliff Myers.

On Wednesday, June 22, 2016 at 5:28:10 PM UTC-6, jim stephens wrote:

"Just keep in mind: you’re never dead if someone remembers you (which can either be a classical cultural reference, or one from as recent as last night)."